Embeddings from Language Models
E771680
deep contextual word representation
natural language processing method
neural network model
word embedding technique
Embeddings from Language Models (ELMo) is a deep contextual word representation technique that uses bidirectional language models to capture rich, context-dependent meanings of words for natural language processing tasks.
All labels observed (1)
| Label | Occurrences |
|---|---|
| Embeddings from Language Models canonical | 1 |
How this entity was disambiguated
This entity first appeared as the object of triple T8993082 — resolving that mention is where its identity was fixed. The disambiguator weighed these candidate entities and picked the highlighted one (or “None”, minting a new entity). This is how homonymy is resolved: the same surface form can point to different entities.
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Embeddings from Language Models Context triple: [Elmo, hasFullName, Embeddings from Language Models]
-
A.
Language Models are Unsupervised Multitask Learners
"Language Models are Unsupervised Multitask Learners" is a 2019 OpenAI research paper that demonstrated how large-scale unsupervised language models like GPT-2 can perform a wide range of tasks without task-specific training.
-
B.
Exploring the Limits of Language Modeling
"Exploring the Limits of Language Modeling" is a research paper that investigates how far large-scale neural language models can be pushed in terms of performance, scalability, and generalization on natural language tasks.
-
C.
Language Models are Few-Shot Learners
"Language Models are Few-Shot Learners" is a landmark research paper that demonstrated large-scale transformer-based language models can perform diverse tasks from just a few examples without task-specific training.
-
D.
LLaMA
LLaMA is a family of large language models developed by Meta AI, designed for efficient training and inference across a range of natural language processing tasks.
-
E.
LLM
LLM is the ICAO airline designator assigned to Yamal Airlines, a Russian regional carrier.
- F. None of above. chosen
- G. Unsure - the case is ambiguous/there is not enough information to decide.
NED2
Entity disambiguation (via description)
gpt-5-mini-2025-08-07
Target entity: Embeddings from Language Models Target entity description: Embeddings from Language Models (ELMo) is a deep contextual word representation technique that uses bidirectional language models to capture rich, context-dependent meanings of words for natural language processing tasks.
-
A.
Language Models are Unsupervised Multitask Learners
"Language Models are Unsupervised Multitask Learners" is a 2019 OpenAI research paper that demonstrated how large-scale unsupervised language models like GPT-2 can perform a wide range of tasks without task-specific training.
-
B.
Exploring the Limits of Language Modeling
"Exploring the Limits of Language Modeling" is a research paper that investigates how far large-scale neural language models can be pushed in terms of performance, scalability, and generalization on natural language tasks.
-
C.
Language Models are Few-Shot Learners
"Language Models are Few-Shot Learners" is a landmark research paper that demonstrated large-scale transformer-based language models can perform diverse tasks from just a few examples without task-specific training.
-
D.
LLaMA
LLaMA is a family of large language models developed by Meta AI, designed for efficient training and inference across a range of natural language processing tasks.
-
E.
LLM
LLM is the ICAO airline designator assigned to Yamal Airlines, a Russian regional carrier.
- F. None of above. chosen
Statements (52)
| Predicate | Object |
|---|---|
| instanceOf |
deep contextual word representation
ⓘ
natural language processing method ⓘ neural network model ⓘ word embedding technique ⓘ |
| basedOn | bidirectional language models ⓘ |
| captures |
context-dependent word meaning
ⓘ
semantic information ⓘ syntactic information ⓘ |
| combinationMethod | learned weighted sum of internal layers ⓘ |
| combines |
backward language model representations
ⓘ
forward language model representations ⓘ |
| comparedTo |
GloVe
ⓘ
linked to:
GloVe word embeddings
word2vec ⓘ |
| developedAt |
Allen Institute for Artificial Intelligence
ⓘ
University of Washington ⓘ |
| developedBy |
Christopher Clark
ⓘ
Kenton Lee ⓘ Luke Zettlemoyer ⓘ Mark Neumann ⓘ Matt Gardner ⓘ Matthew E. Peters ⓘ Mohit Iyyer ⓘ |
| differenceFromStaticEmbeddings | context-dependent representations ⓘ |
| embeddingDimension | 1024 ⓘ |
| hasAbbreviation |
ELMo
ⓘ
linked to:
Deep contextualized word representations
|
| implementedIn | AllenNLP ⓘ |
| improves |
coreference resolution performance
ⓘ
named entity recognition performance ⓘ question answering performance ⓘ semantic role labeling performance ⓘ textual entailment performance ⓘ |
| influenced |
BERT
ⓘ
GPT-style contextual embeddings ⓘ |
| inputRepresentation | character-based ⓘ |
| inputUnit | word ⓘ |
| layerTypes |
character CNN layer
ⓘ
first BiLSTM layer ⓘ second BiLSTM layer ⓘ |
| license | Apache License 2.0 ⓘ |
| numLayers | 3 ⓘ |
| pretrainedOn |
1 Billion Word Benchmark
ⓘ
linked to:
One Billion Word Benchmark
|
| produces | contextualized word embeddings ⓘ |
| publicationTitle | Deep contextualized word representations ⓘ |
| publicationYear | 2018 ⓘ |
| publishedIn | NAACL 2018 ⓘ |
| representationLevel | token-level ⓘ |
| trainingDirection |
backward
ⓘ
forward ⓘ |
| trainingObjective | language modeling ⓘ |
| usage | feature-based transfer learning ⓘ |
| usesArchitecture |
bidirectional LSTM
ⓘ
character-level convolutional neural network ⓘ |
How these facts were elicited
The pipeline generated the facts above by prompting gpt-5.1 with this entity's name + description and the instruction below.
Instruction
You are a knowledge base construction expert. Given a subject entity and a description of it, return factual statements that you know for the subject as a JSON list of dictionaries(triples), where keys must be "subject", "predicate" and "object". The number of facts may be very high, between 25 to 50 or more, for very popular subjects. For less popular subjects, the number of facts can be very low, like 5 or 10. # Requirements - If you don't know the subject at all, return an empty list. - If the subject is not a named entity, return an empty list. - Include at least one triple where predicate is "instanceOf". - Do not get too wordy. - Separate several objects into multiple triples with one object.
Input
Subject: Embeddings from Language Models Description of subject: Embeddings from Language Models (ELMo) is a deep contextual word representation technique that uses bidirectional language models to capture rich, context-dependent meanings of words for natural language processing tasks.
Referenced by (1)
Full triples — surface form annotated when it differs from this entity's canonical label.